WO2022124017A1 - 三次元モデル生成方法及び三次元モデル生成装置 - Google Patents
三次元モデル生成方法及び三次元モデル生成装置 Download PDFInfo
- Publication number
- WO2022124017A1 WO2022124017A1 PCT/JP2021/041943 JP2021041943W WO2022124017A1 WO 2022124017 A1 WO2022124017 A1 WO 2022124017A1 JP 2021041943 W JP2021041943 W JP 2021041943W WO 2022124017 A1 WO2022124017 A1 WO 2022124017A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- dimensional
- points
- line
- image
- dimensional points
- Prior art date
Links
- 238000000034 method Methods 0.000 title claims abstract description 100
- 238000001514 detection method Methods 0.000 claims description 26
- 238000000513 principal component analysis Methods 0.000 claims description 14
- 238000013459 approach Methods 0.000 claims description 8
- 230000005484 gravity Effects 0.000 claims description 8
- 238000012937 correction Methods 0.000 description 113
- 238000010586 diagram Methods 0.000 description 31
- 238000012545 processing Methods 0.000 description 26
- 238000004891 communication Methods 0.000 description 14
- 238000012986 modification Methods 0.000 description 12
- 230000004048 modification Effects 0.000 description 12
- 239000013598 vector Substances 0.000 description 7
- 230000010354 integration Effects 0.000 description 4
- 238000004590 computer program Methods 0.000 description 3
- 238000005516 engineering process Methods 0.000 description 3
- 230000006870 function Effects 0.000 description 3
- 238000003384 imaging method Methods 0.000 description 3
- 230000003287 optical effect Effects 0.000 description 3
- 239000004065 semiconductor Substances 0.000 description 2
- 238000012546 transfer Methods 0.000 description 2
- 238000012935 Averaging Methods 0.000 description 1
- 230000006399 behavior Effects 0.000 description 1
- 238000007796 conventional method Methods 0.000 description 1
- 230000004069 differentiation Effects 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 230000010365 information processing Effects 0.000 description 1
- 230000004807 localization Effects 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 238000013507 mapping Methods 0.000 description 1
- 239000011159 matrix material Substances 0.000 description 1
- 238000005259 measurement Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/50—Depth or shape recovery
- G06T7/55—Depth or shape recovery from multiple images
- G06T7/593—Depth or shape recovery from multiple images from stereo images
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T17/00—Three dimensional [3D] modelling, e.g. data description of 3D objects
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/13—Edge detection
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/70—Determining position or orientation of objects or cameras
- G06T7/73—Determining position or orientation of objects or cameras using feature-based methods
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01C—MEASURING DISTANCES, LEVELS OR BEARINGS; SURVEYING; NAVIGATION; GYROSCOPIC INSTRUMENTS; PHOTOGRAMMETRY OR VIDEOGRAMMETRY
- G01C11/00—Photogrammetry or videogrammetry, e.g. stereogrammetry; Photographic surveying
- G01C11/02—Picture taking arrangements specially adapted for photogrammetry or photographic surveying, e.g. controlling overlapping of pictures
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10004—Still image; Photographic image
- G06T2207/10012—Stereo images
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30244—Camera pose
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N13/00—Stereoscopic video systems; Multi-view video systems; Details thereof
- H04N13/20—Image signal generators
- H04N13/204—Image signal generators using stereoscopic image cameras
- H04N13/239—Image signal generators using stereoscopic image cameras using two 2D image sensors having a relative position equal to or related to the interocular distance
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N13/00—Stereoscopic video systems; Multi-view video systems; Details thereof
- H04N13/20—Image signal generators
- H04N13/204—Image signal generators using stereoscopic image cameras
- H04N13/243—Image signal generators using stereoscopic image cameras using three or more 2D image sensors
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N13/00—Stereoscopic video systems; Multi-view video systems; Details thereof
- H04N2013/0074—Stereoscopic image analysis
- H04N2013/0081—Depth or disparity estimation from stereoscopic image signals
Definitions
- This disclosure relates to a three-dimensional model generation method and a three-dimensional model generation device.
- Patent Document 1 discloses a technique for generating a three-dimensional model of a subject by using a plurality of images obtained by photographing the subject from a plurality of viewpoints.
- the present disclosure provides a three-dimensional model generation method and the like that can improve the generation accuracy of the three-dimensional model.
- the three-dimensional model generation method is a three-dimensional model generation method executed by a computer, in which a first image obtained by photographing a subject from a first viewpoint and the subject are the first. A second image obtained by shooting from two viewpoints is acquired, and at least in the first image, a first line composed of continuous edges and corresponding to the contour of the subject is detected. Based on the first image, the second image, and the first line, a plurality of first three-dimensional points representing the contour are generated in the three-dimensional space in the computer.
- the three-dimensional model generation device includes a computer, and the computer captures a first image obtained by photographing a subject from a first viewpoint and an image of the subject from a second starting point.
- the second image obtained by the above is acquired, and at least in the first image, the first line which is a first line composed of continuous edges and corresponds to the contour of the subject is detected, and the first image is described.
- the second image, and the first line generate a plurality of first three-dimensional points representing the contour in the three-dimensional space in the computer.
- the present disclosure may be realized as a program for causing a computer to execute the steps included in the above-mentioned three-dimensional model generation method. Further, the present disclosure may be realized as a non-temporary recording medium such as a CD-ROM that can be read by a computer that records the program. The disclosure may also be realized as information, data or signals indicating the program. The programs, information, data and signals may be distributed via a communication network such as the Internet.
- the accuracy of three-dimensional model generation can be improved.
- FIG. 1 is a diagram for explaining an outline of a three-dimensional model generation method according to the first embodiment.
- FIG. 2 is a block diagram showing a characteristic configuration of the three-dimensional model generation device according to the first embodiment.
- FIG. 3 is a diagram for explaining the search process according to the first embodiment.
- FIG. 4 is a block diagram showing a functional configuration of the correction unit according to the first embodiment.
- FIG. 5 is a diagram for explaining an example of the process of detecting the contour according to the first embodiment.
- FIG. 6 is a diagram for explaining an example of the process of detecting the contour according to the first embodiment.
- FIG. 7 is a diagram for explaining an example of a process for correcting the positions of a plurality of third three-dimensional points according to the first embodiment.
- FIG. 1 is a diagram for explaining an outline of a three-dimensional model generation method according to the first embodiment.
- FIG. 2 is a block diagram showing a characteristic configuration of the three-dimensional model generation device according to the first embodiment.
- FIG. 8 is a flowchart showing an example of the operation of the three-dimensional model generation device according to the first embodiment.
- FIG. 9 is a flowchart showing an example of details of the generation process of step S104 according to the first embodiment.
- FIG. 10 is a diagram for explaining an example of a process of correcting the positions of a plurality of third three-dimensional points according to the modified example of the first embodiment.
- FIG. 11 is a diagram for explaining an example of a process of correcting the positions of a plurality of third three-dimensional points according to the modified example of the first embodiment.
- FIG. 12 is a diagram for explaining an example of a process of correcting the positions of a plurality of third three-dimensional points according to the modified example of the first embodiment.
- FIG. 10 is a diagram for explaining an example of a process of correcting the positions of a plurality of third three-dimensional points according to the modified example of the first embodiment.
- FIG. 12 is a diagram for explaining an example of a process of correcting the positions of
- FIG. 13 is a diagram for explaining an example of a process of correcting the positions of a plurality of third three-dimensional points according to the modified example of the first embodiment.
- FIG. 14 is a diagram for explaining an example of processing by the correction unit according to the second modification of the first embodiment.
- FIG. 15 is a diagram for explaining an example of processing by the correction unit according to the second modification of the first embodiment.
- FIG. 16 is a block diagram showing a characteristic configuration of the three-dimensional model generation device according to the second embodiment.
- FIG. 17 is a diagram for explaining an example of a process for correcting each frame according to the second embodiment.
- FIG. 18 is a diagram for explaining an example of a process for correcting each frame according to the second embodiment.
- FIG. 19 is a flowchart showing an example of details of the generation process of step S104 according to the second embodiment.
- a three-dimensional model is generated by searching for similarities between a plurality of images.
- search for similarities when searching for similarities in one pixel of one image from another image, epipolar lines on the other image are calculated from the geometric constraints of the camera, and for multiple pixels on the epipolar lines. A search is done. Then, by performing triangulation using the pixels matched in the search for similar points and the position and orientation of the camera when the image was taken, the position of the three-dimensional point shot by the matched pixels is determined. Identify. By doing this for each of the plurality of pixels, a three-dimensional model including a plurality of three-dimensional points is generated.
- the positions of the plurality of three-dimensional points included in the three-dimensional model are affected by the influence of the alias generated in the two-dimensional image.
- the size of the space corresponding to one pixel of the two-dimensional image becomes large, so that the error in the position of the generated three-dimensional point becomes large. Therefore, the accuracy of the generated three-dimensional model is reduced.
- the present disclosure provides a three-dimensional model generation method and the like that can improve the generation accuracy of the three-dimensional model.
- the three-dimensional model generation method is a three-dimensional model generation method executed by a computer, in which a first image obtained by photographing a subject from a first viewpoint and the subject are the first. A second image obtained by shooting from two viewpoints is acquired, and at least in the first image, a first line composed of continuous edges and corresponding to the contour of the subject is detected. Based on the first image, the second image, and the first line, a plurality of first three-dimensional points representing the contour are generated in the three-dimensional space in the computer.
- the first line corresponding to the contour of the subject is detected from the image, and in addition to the plurality of images, a plurality of first three-dimensional points are generated using the detected first line. Since the first line corresponding to the contour of the subject is detected in this way and the first three-dimensional model is generated in consideration of the first line, it is possible to reduce the error due to the alias on the image. Therefore, the accuracy of generating the first three-dimensional model can be improved.
- the first line may be expressed in a unit smaller than the pixel unit in the first image.
- a plurality of second three-dimensional points representing the subject in the three-dimensional space are generated, and the plurality of points are generated.
- a plurality of third-dimensional points corresponding to the contour are identified based on the first line.
- the positions of the plurality of third-dimensional points may be corrected.
- a plurality of third three-dimensional points are a point cloud corresponding to the contour of the subject, and the position of the point cloud is corrected using the specified point cloud. Therefore, the position of the point cloud can be appropriately corrected according to the shape of the contour of the subject.
- a plurality of fourth of the plurality of second three-dimensional points corresponding to the contour to which the second line corresponds based on the second line.
- the position of the uncorrected second three-dimensional point may be corrected.
- a plane is specified by performing a principal component analysis on the plurality of third three-dimensional points, and the plurality of third three-dimensional points are the planes.
- the positions of the plurality of third-dimensional points may be corrected so as to approach.
- the positions of a plurality of third-dimensional points can be accurately corrected according to the shape of the contour.
- the first principal component that passes through the center of gravity of the plurality of third three-dimensional points and is obtained by the principal component analysis is used as a normal line.
- a plane with a small value may be generated as the plane.
- the positions of a plurality of third-dimensional points can be accurately corrected according to the first line which is a line segment.
- an approximate line is generated by performing a minimum square method for the plurality of third three-dimensional points, and the plurality of third three-dimensional points are the approximate lines.
- the positions of the plurality of third-dimensional points may be corrected so as to approach.
- a three-dimensional point whose distance from the first viewpoint is closer than a predetermined distance may be excluded from the plurality of first three-dimensional points.
- the processing load can be reduced without incurring a decrease in accuracy.
- the edge detected from the image having a predetermined resolution or higher among the first image and the second image it is not necessary to use the edge detected from the image having a predetermined resolution or higher among the first image and the second image.
- the processing load can be reduced without causing a decrease in accuracy.
- the detection in the second image, a second line composed of continuous edges and corresponding to the contour of the subject is detected, and in the generation of the plurality of first three-dimensional points, the detection is performed.
- the first image and the second image are corrected by using the first line and the second line, respectively, and (ii) in the corrected first image and the corrected second image. Matching may be performed to search for similarities with each other.
- the first three-dimensional model is generated using the image corrected by using the contour, it is possible to generate the first three-dimensional model in which the error due to the alias on the image is reduced.
- the three-dimensional model generation device includes a computer, and the computer captures a first image obtained by photographing a subject from a first viewpoint and the subject from a second starting point.
- the second image obtained in the above is acquired, and at least in the first image, the first line consisting of continuous edges and corresponding to the contour of the subject is detected, and the first image, the said.
- Based on the second image and the first line a plurality of first three-dimensional points representing the contour are generated in the three-dimensional space in the computer.
- the contour of the subject is detected from the image, and in addition to the plurality of images, the first three-dimensional model is generated using the detected contour. Since the contour of the subject is detected in this way and the first three-dimensional model is generated in consideration of the contour, it is possible to reduce the error due to the alias on the image. Therefore, the accuracy of generating the first three-dimensional model can be improved.
- each figure is a schematic diagram and is not necessarily exactly illustrated. Further, in each figure, the same reference numerals are given to substantially the same configurations, and duplicate explanations may be omitted or simplified.
- FIG. 1 is a diagram for explaining an outline of the three-dimensional model generation method according to the first embodiment.
- FIG. 2 is a block diagram showing a characteristic configuration of the three-dimensional model generation device 100 according to the first embodiment.
- a three-dimensional model of a predetermined region is generated from a plurality of images taken from a plurality of different viewpoints using a plurality of image pickup devices 301.
- the predetermined region is a region including a stationary stationary object, a moving animal body such as a person, or both of them.
- the predetermined region is, for example, a region including at least one of a stationary stationary object and a moving animal body as a subject.
- a predetermined area including a stationary object and an animal body there is a venue where a sports game such as basketball is held, or a space on a road where a person or a car exists.
- the predetermined area may include not only a specific object as a subject but also a landscape or the like.
- FIG. 1 illustrates a case where the subject 500 is a building. Further, in the following, not only a specific object to be a subject but also a predetermined area including a landscape or the like is simply referred to as a subject.
- the three-dimensional model generation system 400 includes an image pickup device group 300 including a plurality of image pickup devices 301, an estimation device 200, and a three-dimensional model generation device 100.
- the plurality of image pickup devices 301 are a plurality of image pickup devices that capture a predetermined area.
- the plurality of image pickup devices 301 each shoot a subject, and the plurality of shot frames are output to the estimation device 200, respectively.
- the image pickup device group 300 includes two or more image pickup devices 301. Further, the plurality of image pickup devices 301 capture the same subject from different viewpoints.
- a frame is, in other words, an image.
- the three-dimensional model generation system 400 is provided with the image pickup device group 300, but the present invention is not limited to this, and one image pickup device 301 may be provided.
- a multi-viewpoint image consisting of a plurality of frames having different viewpoints is generated on one image pickup device 301 while moving one image pickup device 301 for a subject existing in real space. You may let me take a picture.
- Each of the plurality of frames is a frame imaged (generated) by an image pickup device 301 whose position and orientation are different from each other. That is, the plurality of frames are frames taken from different viewpoints.
- each image pickup device 301 may be a camera that generates a two-dimensional image, or may be a camera provided with a three-dimensional measurement sensor that generates a three-dimensional model.
- each of the plurality of image pickup devices 301 is a camera that generates a two-dimensional image by taking a picture.
- the plurality of image pickup devices 301 may be directly connected to the estimation device 200 by wire communication or wireless communication so that the frames captured by each can be output to the estimation device 200, or the communication device, the server, or the like is illustrated. It may be indirectly connected to the estimation device 200 via a hub that does not.
- the frames taken by the plurality of image pickup devices 301 may be output to the estimation device 200 in real time. Also.
- the frame may be recorded once in a memory or an external storage device such as a cloud server, and then output from those external storage devices to the estimation device 200.
- the plurality of image pickup devices 301 may be fixed cameras such as surveillance cameras, mobile cameras such as video cameras, smartphones, or wearable cameras, and movement of drones with a shooting function or the like. It may be a camera.
- the estimation device 200 performs camera calibration by causing one or more image pickup devices 301 to shoot a subject from a plurality of viewpoints.
- the estimation device 200 performs camera calibration for estimating the position and orientation of the plurality of image pickup devices 301 based on the plurality of frames taken by the plurality of image pickup devices 301, for example.
- the posture of the image pickup device 301 indicates at least one of the shooting direction of the image pickup device 301 and the inclination of the image pickup device 301.
- the shooting direction of the image pickup device 301 is the direction of the optical axis of the image pickup device 301.
- the inclination of the image pickup device 301 is the rotation angle of the image pickup device 301 around the optical axis from the reference posture.
- the estimation device 200 estimates the camera parameters of the plurality of image pickup devices 301 based on the plurality of frames acquired from the plurality of image pickup devices 301.
- the camera parameter is a parameter indicating the characteristics of the image pickup device 301, and is an internal parameter including the focal length and the image center of the image pickup device 301, and the position of the image pickup device 301 (more specifically, the three-dimensional position).
- external parameters indicating the posture That is, the position and orientation of each of the plurality of image pickup devices 301 can be obtained by estimating the camera parameters of each of the plurality of image pickup devices 301.
- the estimation method in which the estimation device 200 estimates the position and orientation of the image pickup device 301 is not particularly limited.
- the estimation device 200 may estimate the position and orientation of the plurality of image pickup devices 301 by using, for example, Visual-SLAM (Simultaneus Localization and Mapping) technology.
- the estimation device 200 may estimate the positions and orientations of the plurality of image pickup devices 301 using, for example, the Structure-From-Motion technique.
- FIG. 3 is a diagram for explaining the search process by the restoration unit described later, but since the estimation method also performs a part of the same process, it will be described with reference to FIG.
- the estimation device 200 uses Visual-SLAM technology or Structure-From-Motion technology to extract characteristic points from each of the plurality of frames 531 to 533 photographed by the plurality of image pickup devices 301 as feature points.
- a search for feature points is performed to extract a set of similar similar points among a plurality of frames among the plurality of feature points. Since the estimation device 200 can identify the points on the subject 510 that are commonly reflected in the plurality of frames 531 to 533 by searching for the feature points, the estimation device 200 can use the extracted set of similar points on the subject 510.
- the three-dimensional coordinates (that is, the three-dimensional position) of the point can be obtained by the principle of triangulation.
- the estimation device 200 can estimate the position and orientation of each image pickup device 301 by extracting a plurality of sets of similar points and using the plurality of sets of similar points.
- the estimation device 200 calculates three-dimensional coordinates for each set of similar points in the process of estimating the position and orientation of each image pickup device 301, and the estimation device 200 is a tertiary including a plurality of three-dimensional points indicated by the calculated plurality of three-dimensional coordinates.
- the original model 520 may be generated.
- Each of the plurality of three-dimensional points indicates a position on the subject in the three-dimensional space.
- the estimation device 200 obtains the position and orientation of each image pickup device 301 and the map information as the estimation result.
- the three-dimensional model 520 includes the three-dimensional positions of each of the plurality of three-dimensional points.
- the three-dimensional model 520 not only displays a plurality of three-dimensional positions, but also displays the color of each three-dimensional point, the surface shape around each three-dimensional point, and information indicating which frame each three-dimensional point was generated by. It may be included.
- the estimation device 200 may generate a three-dimensional model 520 including a sparse three-dimensional point cloud by limiting the number of sets of similar points to a predetermined number in order to speed up the estimation process. .. This is because the estimation device 200 can estimate the position and orientation of each image pickup device 301 with sufficient accuracy even if the set of similarities is limited to a predetermined number.
- the predetermined number may be determined to be a number capable of estimating the position and orientation of each image pickup device 301 with sufficient accuracy.
- the estimation device 200 may estimate the position and orientation of each image pickup device 301 by using a set of similar points that are similar to each other with a predetermined degree of similarity or higher. As a result, the estimation device 200 can limit the number of sets of similarities used in the estimation process to the number of sets that are similar at a predetermined degree of similarity or higher.
- the estimation device 200 may calculate, for example, the distance between the image pickup device 301 and the subject 510 as a camera parameter based on the position and orientation of the image pickup device 301 estimated by using the above technique.
- the three-dimensional model generation system 400 includes a distance measuring sensor, and the distance between the image pickup apparatus 301 and the subject 510 may be measured using the distance measuring sensor.
- the estimation device 200 may be directly connected to the three-dimensional model generation device 100 by wire communication or wireless communication, or may be indirectly connected to the estimation device 200 via a hub (not shown) such as a communication device or a server. You may. As a result, the estimation device 200 outputs the plurality of frames received from the plurality of image pickup devices 301 and the plurality of camera parameters of the plurality of estimated image pickup devices 301 to the three-dimensional model generation device 100.
- the plurality of camera parameters estimated by the estimation device 200 may be output to the three-dimensional model generation device 100 in real time. Further, the plurality of camera parameters may be once recorded in an external storage device such as a memory or a cloud server, and then output from those external storage devices to the three-dimensional model generation device 100.
- an external storage device such as a memory or a cloud server
- the estimation device 200 is a computer including, for example, a control program, a processing circuit such as a processor or a logic circuit that executes the control program, and a recording device such as an internal memory for storing the control program or an accessible external memory. At least have a system.
- the three-dimensional model generation device 100 generates a three-dimensional model of a predetermined region based on a plurality of frames taken by the plurality of image pickup devices 301 and camera parameters estimated by the estimation device 200. Specifically, the three-dimensional model generation device 100 puts a three-dimensional model of a subject on a virtual three-dimensional space based on a plurality of camera parameters corresponding to each of the plurality of image pickup devices 301 and a plurality of frames. It is a device that executes the 3D model generation process to be generated.
- the three-dimensional model of the subject is data including the three-dimensional shape of the subject and the color of the subject, which are restored on a virtual three-dimensional space from the frame in which the actual subject is photographed.
- the three-dimensional model of the subject is a multi-viewpoint, that is, a point indicating the three-dimensional position of each of the plurality of points on the subject captured in each of the plurality of two-dimensional images taken by the plurality of image pickup devices 301 at a plurality of different viewpoints. It is a set.
- the three-dimensional position is represented by, for example, ternary information including an X component, a Y component, and a Z component indicating the positions of the X axis, the Y axis, and the Z axis that are orthogonal to each other. That is, the three-dimensional position is represented by three-dimensional coordinates in the three-dimensional Cartesian coordinate system.
- the three-dimensional position is not limited to being represented by the three-dimensional coordinates in the three-dimensional Cartesian coordinate system, but may be represented by the three-dimensional coordinates in the three-dimensional polar coordinate system.
- the information including the plurality of points indicating the three-dimensional position includes not only the three-dimensional position (that is, the information indicating the coordinates) but also the information indicating the color of each point. Attribute information or the like representing the surface shape of each point and its surroundings may be included.
- the three-dimensional model generation device 100 includes, for example, a control program, a processing circuit such as a processor or a logic circuit that executes the control program, and a recording device such as an internal memory for storing the control program or an accessible external memory. At least equipped with a computer system.
- the three-dimensional model generation device 100 is an information processing device. The functions of each processing unit of the three-dimensional model generator 100 may be realized by software or hardware.
- the three-dimensional model generation device 100 may store the camera parameters of the plurality of image pickup devices 301 in advance. In this case, the three-dimensional model generation system 400 does not have to include the estimation device 200. Further, the plurality of image pickup devices 301 may be wirelessly or wiredly connected to the three-dimensional model generation device 100 so as to be communicable.
- the plurality of frames taken by the image pickup apparatus 301 may be directly output to the three-dimensional model generation apparatus 100.
- the image pickup device 301 may be directly connected to the three-dimensional model generation device 100 by, for example, wired communication or wireless communication, or the three-dimensional model generation may be performed via a hub (not shown) such as a communication device or a server. It may be indirectly connected to the device 100.
- the three-dimensional model generation device 100 is a device that generates a three-dimensional model from a plurality of frames.
- the three-dimensional model generation device 100 includes a reception unit 110, a storage unit 120, an acquisition unit 130, a generation unit 140, and an output unit 150.
- the receiving unit 110 receives from the estimation device 200 a plurality of frames taken by the plurality of image pickup devices 301 and a plurality of camera parameters estimated by the estimation device 200.
- the receiving unit 110 has a first frame (first image) obtained by shooting the subject from the first viewpoint and a second frame (first image) obtained by shooting the subject from the second viewpoint. 2 image) and is acquired. That is, the plurality of frames received by the receiving unit 110 include the first frame and the second frame.
- the plurality of frames are, for example, a plurality of frames obtained by photographing the subject from a plurality of different viewpoints when the subject is a stationary object.
- the plurality of frames are, for example, a plurality of frames obtained by photographing the subject at a predetermined timing (that is, at the same time) from a plurality of viewpoints when the subject is an animal body.
- the three-dimensional model generation device 100 performs a process of generating a first three-dimensional model using a frame set having a plurality of frames as one set as a processing unit. Framesets are sometimes referred to as multi-viewpoint images.
- the receiving unit 110 may acquire the three-dimensional model 520 from the estimation device 200.
- the receiving unit 110 outputs a plurality of received frames and camera parameters to the storage unit 120.
- the receiving unit 110 is, for example, a communication interface for communicating with the estimation device 200.
- the receiving unit 110 includes, for example, an antenna and a wireless communication circuit.
- the receiving unit 110 includes, for example, a connector connected to a communication line and a wire communication circuit.
- the receiving unit 110 may receive a plurality of frames from a plurality of image pickup devices 301 without going through the estimation device 200.
- the storage unit 120 stores a plurality of frames and a plurality of camera parameters received by the reception unit 110.
- the storage unit 120 may store the three-dimensional model 520 received by the reception unit 110.
- the storage unit 120 may store the processing results of each processing unit included in the three-dimensional model generation device 100.
- the storage unit 120 stores, for example, a control program executed by each processing unit included in the three-dimensional model generation device 100.
- the storage unit 120 is realized by, for example, an HDD (Hard Disk Drive), a flash memory, or the like.
- the acquisition unit 130 acquires a plurality of frames stored in the storage unit 120 and the camera parameters of each image pickup device 301 from the storage unit 120 and outputs them to the generation unit 140.
- the three-dimensional model generation device 100 does not have to include the storage unit 120 and the acquisition unit 130.
- the receiving unit 110 may output the plurality of frames received from the plurality of image pickup devices 301 and the camera parameters of each image pickup device 301 received from the estimation device 200 to the generation unit 140.
- the generation unit 140 generates a first three-dimensional model using a plurality of frames and a plurality of camera parameters.
- the first three-dimensional model contains a plurality of first three-dimensional points.
- the generation unit 140 has a restoration unit 141 and a correction unit 142.
- the restoration unit 141 generates a plurality of second three-dimensional points representing a subject in a three-dimensional space by performing matching for searching for similarities between the plurality of frames, for example, the first frame and the second frame. ..
- the correction unit 142 corrects the positions of the plurality of second three-dimensional points. As a result, the first three-dimensional model is generated.
- the plurality of first three-dimensional points included in the first three-dimensional model are generated by correcting at least a part of the plurality of second three-dimensional points by the correction unit 142.
- the restoration unit 141 and the correction unit 142 will be specifically described.
- the restoration unit 141 searches for similarities similar to the first point of the first frame among the plurality of frames from a plurality of second points in the search area based on the first point in the second frame.
- the first frame is a frame (image) that is a reference source for the first point among the plurality of frames.
- the second frame is one of the remaining plurality of frames excluding the first frame from the plurality of frames, and is a frame (image) to which the search for similarities similar to the first point is referred. be.
- the first point is specifically the first pixel of one of the plurality of first pixels.
- the plurality of first pixels are pixels included in the first frame, for example, all pixels included in the first frame.
- the search region is a region defined by an epipolar line corresponding to the first point of the first frame in a plurality of second frames, and is, for example, a region composed of a plurality of second points on the epipolar line. ..
- the plurality of second points are, specifically, a plurality of second pixels included in the search area.
- the restoration unit 141 may search for similar points (similar pixels) similar to the first pixel from the plurality of second pixels included in the second frame for each of the plurality of first pixels.
- the plurality of second pixels are, for example, all the pixels included in the second frame.
- the restoration unit 141 may search for similarities of the first pixel by using each of the plurality of second frames as a search target, or search at least one second frame of the plurality of second frames as a search target. , The search for the similarity of the first pixel may be performed.
- FIG. 3 is a diagram for explaining the search process by the restoration unit 141.
- the image pickup device 301 of the first viewpoint V1 takes a picture of the first frame 531 including the subject 510
- the image pickup device 301 of the second viewpoint V2 takes a picture of the second frame 532 including the subject 510
- the third viewpoint V3 An example is shown in which the second frame 533 including the subject 510 is photographed by the image pickup apparatus 301 of the above.
- a straight line connecting the position of the image pickup apparatus 301 that captured the first frame and the two-dimensional coordinates on the first frame of the first pixel for each first pixel becomes the second frame to be processed.
- Calculate the projected epipolar line For example, as shown in FIG. 3, the restoration unit 141 calculates the epipolar line 552 projected on the second frame 532 by the straight line L1 connecting the first viewpoint V1 and the first pixel 541. Further, the restoration unit 141 calculates the epipolar line 553 in which the straight line L1 is projected on the second frame 533. Then, the restoration unit 141 searches for similarities similar to the first pixel 541 to be processed in the first frame 531 from the epipolar lines 552 and 552, respectively.
- the restoration unit 141 calculates the degree of similarity between the first pixel and each of the plurality of second pixels in the search area. Specifically, the restoration unit 141 formulates Equation 1 with N (I, J) indicating a Normalized Cross Correlation (NCC) between a small area between the first frame and the second frame to be processed as the similarity. Calculate using. N (I, J) is represented by a numerical value between -1 and 1, and the closer to 1 is, the higher the similarity is.
- the restoration unit 141 may calculate SSD (I, J) as the degree of similarity instead of NCC.
- SSD Sud of Squared Difference
- SSD (I, J) indicates that the smaller the value, the higher the similarity.
- the restoration unit 141 identifies the second pixels 542 and 543, which have the highest degree of similarity with the first pixel 541, as similarities to the first pixel.
- the restoration unit 141 uses the first pixel 541 and the second pixels 542 and 543 selected as similar points to the first pixel 541 to triangulate a three-dimensional point indicated by the three-dimensional coordinates of the point on the subject 510. It is calculated by the principle of.
- the restoration unit 141 calculates one three-dimensional point corresponding to one first pixel by calculating the average of the plurality of three-dimensional points. May be generated.
- the restoration unit 141 will add the camera identification information of the image pickup device 301 that generated each frame to the header of each frame. May be added, or a plurality of camera specific information of the plurality of image pickup devices 301 may be added to the header of the frame set including the plurality of frames.
- the camera specific information may include identification information that identifies the image pickup device 301 to which the camera specific information corresponds, or may include camera parameters of the image pickup device 301 to which the camera specific information corresponds.
- FIG. 4 is a block diagram showing the functional configuration of the correction unit 142.
- the correction unit 142 has a detection unit 142a and a three-dimensional point correction unit 142b.
- the detection unit 142a detects a line consisting of continuous edges and corresponding to the contour of the subject 510 at least in the first frame.
- the line detected in the first frame is an example of the first line.
- the detection unit 142a may detect an edge by using a gradient method using a first derivative with respect to the pixel value of each pixel of the first frame. Further, the detection unit 142a may detect the edge by using the Laplacian method using the second derivative with respect to the pixel value of each pixel of the first frame.
- the edge detected by the detection unit 142a is a point (edge point) in the first frame.
- the detection unit 142a detects continuous edges by using a plurality of edges detected on the first frame.
- a well-known algorithm can be used for the detection unit 142a to detect continuous edges.
- the detection unit 142a may detect continuous linear edges by using, for example, an LSD (Line Segment Detector). By using LSD, the detection unit 142a detects the start point and the end point of the continuous edge, and detects the straight line connecting the start point and the end point as the continuous edge.
- LSD Line Segment Detector
- the contours corresponding to the lines detected by the detection unit 142a are, for example, line segments, curves, planes, curved surfaces, and the like.
- the contour may be a line or surface that forms part of a particular shape.
- the specific shape may be a prism, a pyramid, a cylinder, a cone, a sphere, or the like, or a three-dimensional object formed by rotating a line segment or a curve around a certain axis of rotation, that is, a rotating body. May be good.
- the contour may be distinguished for each type of line shape, for example.
- the shape of the line may be a straight line (line segment) connecting two points, or a curve connecting two points.
- the curve connecting the two points may be a smooth curve having no bent portion.
- the line detected by the detection unit 142a may be a line having a finite length separated by two points.
- the contour may be a surface including the detected line.
- the contour may be a plane when the shape of the detected line is a line segment, or may be a curved surface when the shape of the detected line is a curve.
- 5 and 6 are diagrams for explaining an example of the process of detecting a line by the detection unit 142a.
- the detection unit 142a detects a line segment.
- the region 561 including the linear shape of the subject 510 is shown by a broken line. Since the first frame 531 expresses the subject 510 in pixel units, the contour of the subject 510 is also expressed in pixel units. Therefore, in general, in the region 561 in which the linear shape extending in the diagonal direction of the first frame 531 is photographed, as shown in FIG. 6A, a large number of aliases in which the edges look like steps are generated. However, since the detected continuous edges are not represented in pixel units, the contour of the subject 510 is represented by the continuous edges with high reproducibility.
- the position of the detected continuous edge is not limited to the position of the pixel because it is specified by the first-order or second-order differentiation to the brightness gradient obtained from the relationship between the position of each pixel and the pixel value. It is specified by the accuracy (minor pixel accuracy) in the area unit smaller than the pixel unit. That is, the detection unit 142a detects a line consisting of continuous edges at a position with a decimal pixel accuracy. A line consisting of continuous edges is represented by a unit smaller than a pixel unit in the first frame 531. Therefore, the continuous edge expresses the contour 562 in a region unit smaller than the pixel unit, which has high reproducibility. Further, since the detection unit 142a detects continuous edges by using the first derivative or the second derivative to the luminance gradient, it is possible to obtain the result in which the folding distortion (alias) in the plurality of pixels is offset.
- 3D points instead of generating 3D points only from the pixels of the 1st frame 531 by using continuous edges that can express the contour of the subject 510 with high reproducibility, 3D points are generated. It is possible to generate more accurate 3D points.
- the three-dimensional point correction unit 142b identifies a group including a plurality of third-dimensional points included in the contour corresponding to the line 562 among the plurality of second-three-dimensional points.
- the group may include a plurality of three-dimensional points generated from a plurality of pixels through which the line 562 passes in the first frame as a plurality of third three-dimensional points.
- the group may include a plurality of three-dimensional points generated from a plurality of pixels through which the line 562 passes in the first frame and pixels around the plurality of pixels as a plurality of third three-dimensional points. ..
- the pixels around the plurality of pixels may include all of the plurality of pixels located closer to the pixels separated by a predetermined number from the plurality of pixels.
- the three-dimensional point correction unit 142b specifies a group including a plurality of third-dimensional points specified in the portion corresponding to the line for each of the one or more lines detected by the detection unit 142a. ..
- the three-dimensional point correction unit 142b corrects the positions of a plurality of third three-dimensional points belonging to the group specified by the line based on each line. Specifically, the three-dimensional point correction unit 142b identifies a plane by performing principal component analysis on a plurality of third three-dimensional points belonging to each group for each group, and the three-dimensional point correction unit 142b identifies a plane, and a plurality of third three-dimensional points. Corrects the positions of the plurality of third-dimensional points so that is closer to the plane. In this way, the three-dimensional point correction unit 142b corrects the positions of the plurality of third-dimensional points corresponding to the contours among the plurality of second-three-dimensional points. As a result, a plurality of first three-dimensional points are generated.
- FIG. 7 is a diagram for explaining an example of a process of correcting the positions of a plurality of third three-dimensional points by the three-dimensional point correction unit 142b according to the first embodiment.
- FIG. 7A shows a plurality of third three-dimensional points 570 belonging to one group.
- the three-dimensional point correction unit 142b calculates the first principal component 571 and the second principal component 572 as shown in FIG. 7 (b) by performing principal component analysis on a plurality of third three-dimensional points 570. do. Specifically, the three-dimensional point correction unit 142b calculates the variance-covariance matrix of the three-dimensional coordinates of the plurality of third three-dimensional points 570, and the obtained principal component (eigenvector) has the largest magnitude in order. The two selected principal components are calculated as the first principal component 571 and the second principal component 572.
- the three-dimensional point correction unit 142b generates a first plane based on the first principal component 571 and a second plane based on the second principal component 572 as candidate planes (candidate planes) to be used for correction.
- the first plane is a plane that passes through the centers of gravity of the plurality of third three-dimensional points 570 and has the first principal component 571 as a normal.
- the second plane is a plane that passes through the centers of gravity of the plurality of third three-dimensional points 570 and has the second main component 572 as a normal.
- the three-dimensional point correction unit 142b determines, among the first plane and the second plane, a plane having a small total distance from a plurality of third three-dimensional points 570 as a plane to be used for correction.
- the equation of the plane showing the candidate plane is expressed by the equation 3.
- Equation 4 the distance D between this plane and the point (x0, y0, z0) is expressed by Equation 4.
- Equation 5 The sum of the distances, Dollar, is expressed by Equation 5.
- the three-dimensional point correction unit 142b calculates the Dollar for each of the first plane and the second plane, and determines the plane having the minimum Dollar as the correction plane 573. As a result, the three-dimensional point correction unit 142b determines the first plane based on the first principal component 571 as the correction surface 573 in the example shown in FIG. 7 (c).
- the three-dimensional point correction unit 142b determines the first frame corresponding to the third three-dimensional point for each of the plurality of third three-dimensional points 570, as shown in FIG. 7 (d). The intersection of the projection line of the upper pixel on the third three-dimensional point and the correction surface 573 is calculated as the correction point 574.
- the third three-dimensional point (Xn, Yn, Zn) corresponding to the pixel t ( It can be expressed as Vx, Vy, Vz).
- the three-dimensional unit vector (Vx, Vy, Vz) corresponding to the pixel is expressed by the following equation 8 assuming a pinhole camera.
- Equation 8 (x, y) are the two-dimensional coordinates of the pixels in the first frame.
- (Cx, cy) is the center position of the first frame in the internal parameters of the image pickup apparatus 301 in which the first frame is photographed.
- (Fx, fy) is the focal length in the internal parameter of the image pickup apparatus 301 in which the first frame is photographed.
- the three-dimensional point correction unit 142b corrects the positions of the plurality of third three-dimensional points 570 to generate a first three-dimensional model including the plurality of correction points 574.
- the plurality of correction points 574 are aligned along the line detected in the first frame. Therefore, in the generated first three-dimensional model, when viewed from the viewpoint of the first frame, the portion corresponding to the line detected in the first frame has a shape along the line.
- the plurality of first three-dimensional points are generated as a result of the correction of the plurality of third three-dimensional points 570, the plurality of correction points 574 are included. That is, the plurality of first three-dimensional points include a plurality of uncorrected second three-dimensional points and a plurality of correction points 574 among the plurality of second three-dimensional points.
- FIG. 8 is a flowchart showing an example of the operation of the three-dimensional model generation device 100.
- the receiving unit 110 receives from the estimation device 200 a plurality of frames taken by the plurality of image pickup devices 301 and the camera parameters of each image pickup device 301 (S101).
- Step S101 is an example of a step of acquiring a plurality of images.
- the receiving unit 110 does not have to receive the plurality of frames and the plurality of camera parameters at one timing, and may receive each at different timings. That is, the acquisition of a plurality of frames and the acquisition of a plurality of camera parameters may be performed at the same timing or at different timings from each other.
- the storage unit 120 stores a plurality of frames taken by the plurality of image pickup devices 301 received by the reception unit 110 and the camera parameters of each image pickup device 301 (S102).
- the acquisition unit 130 acquires a plurality of frames and a plurality of camera parameters stored in the storage unit 120, and outputs the acquired plurality of frames and a plurality of camera parameters to the generation unit 140 (S103). ..
- the generation unit 140 generates a first three-dimensional model using a plurality of frames and a plurality of camera parameters (S104).
- the details of step S104 for generating the first three-dimensional model will be described later with reference to FIG.
- the output unit 150 outputs the first three-dimensional model generated by the generation unit 140 (S105).
- FIG. 9 is a flowchart showing an example of details of the generation process of step S104 by the generation unit 140.
- the generation unit 140 performs loop 1 for each frame set of multi-viewpoint images taken at timings corresponding to each other (S111). In loop 1, loop 2 is performed for each frame set.
- the generation unit 140 performs loop 2 for each of a plurality of first pixels of the first frame in the frame set to be processed (S112). In the loop 2, the processes of steps S113 to S116 are performed for each of the plurality of first pixels.
- the restoration unit 141 transfers the first pixel to be processed from the plurality of second pixels on the epipolar line corresponding to the first pixel on the plurality of second frames of the frame set to be processed to the first pixel. Search for similar similarities (S113).
- the restoration unit 141 generates a second three-dimensional point by performing a triangulation survey using the search result obtained in step S113. As a result, the restoration unit 141 generates a second three-dimensional model including a plurality of second three-dimensional points (S114). Details of steps S113 and S114 are omitted because they are described in the description of the restoration unit 141.
- the correction unit 142 detects a line consisting of edges continuous from the first frame and corresponding to the contour of the subject (S115).
- the correction unit 142 corrects the positions of the plurality of third-dimensional points corresponding to the detected lines among the plurality of second-three-dimensional points according to the shape of the line (S116). As a result, the correction unit 142 generates a first three-dimensional model including a plurality of first three-dimensional points. Details of steps S115 and S116 are omitted because they are described in the description of the correction unit 142.
- Loop 2 ends when the processing of steps S113 to S116 is completed for all the first pixels included in the first frame of the frame set to be processed.
- Loop 1 ends when loop 2 ends for all framesets.
- the loop 2 includes both a process of generating a second three-dimensional model and a process of correcting the positions of a plurality of third three-dimensional points. Not limited to, these two processes may not be included in the same loop. That is, the loop of the process of generating the second three-dimensional model and the loop of the process of correcting the positions of the plurality of third three-dimensional points may be independent.
- the first frame obtained by shooting the subject from the first viewpoint and the second frame obtained by shooting the subject from the second viewpoint are acquired.
- S101 at least in the first frame, a line consisting of continuous edges and corresponding to the contour of the subject is detected (S115), and based on the first frame, the second frame, and the line, Generate a plurality of first three-dimensional points representing contours in a three-dimensional space in a computer (S116).
- a line corresponding to the contour of the subject is detected from the frame, and in addition to the plurality of frames, a plurality of first 3D points are generated using the detected line.
- the line corresponding to the contour of the subject is detected and a plurality of first three-dimensional points are generated in consideration of the line, it is possible to reduce the error due to the alias on the image. Therefore, the accuracy of generating the first three-dimensional model can be improved.
- the contour represented by the plurality of first three-dimensional points in the three-dimensional space in the computer when viewed from the viewpoint corresponding to the first viewpoint is a line. It has a shape along it. That is, in the three-dimensional model generation method, since a plurality of first three-dimensional points are generated along the shape of the contour of the subject when viewed from the first viewpoint, the generation accuracy of the first three-dimensional model is improved. Can be done.
- a plurality of subjects are expressed in a three-dimensional space in a computer by performing matching for searching for similarities between the first frame and the second frame. (S113, S114), and for the generation of a plurality of 1st 3D points (S116), the contour corresponds to the contour of the plurality of 2D points based on the line. A plurality of 3D points are specified, and the positions of the plurality of 3D points are corrected.
- a plurality of third three-dimensional points are a point cloud corresponding to the contour, and the position of the point cloud is corrected using the specified point cloud. Therefore, the position of the point cloud can be appropriately corrected according to the shape of the contour.
- a plane is formed by performing a principal component analysis on a plurality of third three-dimensional points. It is generated and the positions of the plurality of third-dimensional points are corrected so that the plurality of third-dimensional points approach the plane. Therefore, the positions of the plurality of third three-dimensional points can be accurately corrected according to the shape of the contour.
- the center of gravity of the plurality of third three-dimensional points is passed and the principal component analysis is performed.
- a plane having a small total distance from the third three-dimensional point is generated as a plane used for correction. Therefore, even if the length of the detected line is not sufficiently long, the positions of the plurality of third-dimensional points can be accurately corrected according to the line segment.
- Modification example (Modification 1)
- the three-dimensional model generation device 100 performs principal component analysis on a plurality of third three-dimensional points belonging to the group specified by the line to correct the surface. Is generated, and the positions of a plurality of third-dimensional points are corrected using the correction surface, but the present invention is not limited to this.
- an approximate line is generated by performing a least squares method on a plurality of third three-dimensional points, and the positions of the plurality of third three-dimensional points are determined using the approximate line. It may be corrected.
- the three-dimensional point correction unit 142b identifies a group including a plurality of third three-dimensional points included in the portion corresponding to the line 562 among the plurality of second three-dimensional points. Then, the three-dimensional point correction unit 142b generates an approximate line by performing a least squares method on a plurality of third three-dimensional points. The three-dimensional point correction unit 142b corrects the positions of the plurality of third-dimensional points so that the plurality of third-dimensional points approach the approximate line.
- FIG. 10 to 13 are diagrams for explaining an example of a process of correcting the positions of a plurality of third three-dimensional points by the three-dimensional point correction unit 142b according to the modified example of the first embodiment.
- FIG. 10 shows a plurality of third three-dimensional points 580 belonging to one group.
- the three-dimensional point correction unit 142b projects a plurality of third three-dimensional points 580 onto the XZ plane.
- the number of a plurality of third three-dimensional points 580 belonging to the group is n, and the coordinates of the i-th three-dimensional point in the XZ plane are (Xi, Zi). Then, it is assumed that the formula of the approximate line to be obtained is shown by the formula 9.
- the three-dimensional point correction unit 142b can calculate a and b by the least squares method using the following equation 10.
- the three-dimensional point correction unit 142b has a plurality of third three-dimensional points in the XZ plane when the two-dimensional plane in which the plurality of third three-dimensional points 580 exist is the XZ plane.
- the least squares method for 580 it is possible to generate approximation lines 581 for a plurality of third three-dimensional points 580.
- the three-dimensional point correction unit 142b calculates the intersection with the approximation line 581 for each third three-dimensional point.
- the three-dimensional point correction unit 142b calculates the Y coordinate of the correction point on the XYZ coordinates by using the calculated coordinates of the intersection and the equation 12.
- the three-dimensional unit vector corresponding to the pixel corresponding to the i-th third three-dimensional point is represented by the equation 14.
- the three-dimensional point correction unit 142b samples two points from a plurality of third three-dimensional points to calculate a plane 582.
- the normal of plane 582 is represented by equation 15 and the plane is represented by equation 16.
- Equation 16 The normal can be obtained by the outer product of the vectors as shown in Equation 17.
- the three-dimensional point correction unit 142b corrects the positions of the plurality of third three-dimensional points 580 to generate a first three-dimensional model including the plurality of correction points 583.
- the plurality of correction points 583 are aligned along the line detected in the first frame. Therefore, in the generated first three-dimensional model, the contour represented by the plurality of first three-dimensional points in the three-dimensional space in the computer when viewed from the starting point corresponding to the first viewpoint has a shape along a line.
- the plurality of correction points 583 are included. That is, the plurality of first three-dimensional points include a plurality of uncorrected second three-dimensional points and a plurality of correction points 583 among the plurality of second three-dimensional points.
- an approximate line is generated by performing a least squares method on a plurality of third three-dimensional points.
- the positions of the plurality of third-dimensional points are corrected so that the plurality of third-dimensional points approach the approximate line. Therefore, it is possible to accurately correct the positions of a plurality of third-dimensional points according to the approximate lines that are close to the point cloud.
- the three-dimensional model generation device 100 may generate a surface for correction by using a plurality of third-dimensional points that have already been corrected. In this case, since the processing of the correction unit 142 is different, the processing of the correction unit 142 will be described.
- FIG. 14 is a diagram for explaining an example of processing by the correction unit 142 according to the modification 2 of the first embodiment.
- FIG. 15 is a diagram for explaining an example of processing by the correction unit 142 according to the second modification of the first embodiment.
- the detection unit 142a of the correction unit 142 detects the line 601 consisting of continuous edges in the first frame 591 and corresponds to the contour of the subject. After that, the detection unit 142a detects in the second frame 592 a line 602 composed of continuous edges and corresponding to the contour of the subject.
- the line 602 corresponds to a contour including at least a part of the contour of the subject to which the line 601 corresponds. That is, the contour corresponding to the line 602 and the contour corresponding to the line 601 include a portion overlapping with each other.
- the line 601 corresponds to the first line, and the line 602 corresponds to the second line.
- the detection unit 142a detects a line corresponding to the contour of the subject in the second frame 592, compares the detected line with the line 601 and matches the detected line with the line 601.
- a line whose degree is higher than a predetermined degree of coincidence is detected as line 602.
- the degree of coincidence between the detected line and the line 601 may be, for example, the degree of similarity between the normal vector of the detected line and the normal vector of the line 601.
- the degree of coincidence is calculated based on the similarity between the normal vectors and the distance between the detected line in the three-dimensional space and the position of the line 601 in the three-dimensional space. You may. In this case, the degree of matching is calculated so that the shorter the distance, the greater the degree of matching. This similarity can be calculated using a well-known method.
- the three-dimensional point correction unit 142b identifies a first group including a plurality of third three-dimensional points 610 included in the contour corresponding to the line 601 among the plurality of second three-dimensional points. As described above, the three-dimensional point correction unit 142b has a plurality of third-dimensional dimensions specified by the portion corresponding to the line 601 for each of the one or more lines 601 detected in the first frame 591 by the detection unit 142a. Identify a first group that includes point 610.
- the three-dimensional point correction unit 142b corrects the positions of the plurality of third three-dimensional points 610 belonging to the first group specified by the line 601 based on each line 601. Specifically, the three-dimensional point correction unit 142b identifies the plane 621 by performing principal component analysis on a plurality of third three-dimensional points 601 belonging to each first group for each first group, and a plurality of planes 621 are specified. The positions of the plurality of third-dimensional points 610 are corrected so that the third three-dimensional point 601 of the above is closer to the plane 621. In this way, the three-dimensional point correction unit 142b corrects the positions of the plurality of third three-dimensional points 610 corresponding to the contour among the plurality of second three-dimensional points. As a result, a plurality of first three-dimensional points 611 are generated.
- the three-dimensional point correction unit 142b identifies a plurality of fourth three-dimensional points 613 included in the contour corresponding to the line 602 among the plurality of second three-dimensional points.
- the plurality of fourth three-dimensional points 613 include a plurality of corrected third three-dimensional points 611 and two uncorrected second three-dimensional points 612.
- the three-dimensional point correction unit 142b has a plurality of fourth and third dimensions specified by the portion corresponding to the line 602 for each of the one or more lines 602 detected in the second frame 592 by the detection unit 142a. Identify a second group that includes point 613.
- the three-dimensional point correction unit 142b corrects the positions of the plurality of fourth three-dimensional points 613 belonging to the second group specified by the line 602 based on each line 602. Specifically, the three-dimensional point correction unit 142b corrects the position of at least the uncorrected second three-dimensional point 612 among the plurality of fourth three-dimensional points 613.
- the three-dimensional point correction unit 142b identifies the plane 622 by performing principal component analysis on a plurality of fourth three-dimensional points 613 belonging to each second group for each second group, and a plurality of fourth tertiary points. The positions of the plurality of fourth three-dimensional points 613 are corrected so that the original point 613 approaches the specified plane 622.
- the three-dimensional point correction unit 142b may correct the positions of all the plurality of fourth three-dimensional points 613, or among the plurality of fourth three-dimensional points 613, the uncorrected second three-dimensional point. Only the position of 612 may be corrected.
- the distance D1 from the first viewpoint V11 to the contour of the subject corresponding to the line 601 detected in the first frame 591 was detected in the second frame 592 from the second viewpoint V12. If the distance to the contour of the subject corresponding to line 602 is shorter than D2, a plurality of 4th and 3D dimensions are used by using a plane on an extension of the specified plane 621 to correct the plurality of 3D points 610. Of the points 613, at least the uncorrected second three-dimensional point 612 may be corrected.
- a plurality of three-dimensional points 610 that have already been corrected are also used to correct at least the uncorrected second three-dimensional point 612, so that the subject is continuous. It is possible to accurately generate the positions of a plurality of first three-dimensional points corresponding to the contour.
- FIG. 16 is a block diagram showing a characteristic configuration of the three-dimensional model generation device 100A according to the second embodiment.
- the configuration of the three-dimensional model generation device 100A according to the second embodiment is different from that of the three-dimensional model generation device 100 according to the first embodiment. Therefore, the generation unit 140A will be described. Since the components other than the generation unit 140A of the three-dimensional model generation device 100A are the same as the components of the three-dimensional model generation device 100 of the first embodiment, the same reference numerals are given and the description thereof will be omitted.
- the generation unit 140A is different from the generation unit 140 in that after correcting a plurality of frames, the first three-dimensional model is generated by using the corrected plurality of frames.
- the generation unit 140A has a correction unit 141A and a restoration unit 142A.
- the correction unit 141A detects a line consisting of continuous edges from each of the plurality of frames and corresponding to the contour of the subject. That is, the correction unit 141A detects the first line consisting of continuous edges in the first frame. Further, the correction unit 141A detects a second line composed of continuous edges in the second frame. This process is the same as the process of detecting a line from the frame of the correction unit 142 according to the first embodiment.
- the correction unit 141A corrects each of the plurality of frames by using the lines detected in each of the plurality of frames. That is, the correction unit 141A corrects each of the first frame and the second frame by using the first line and the second line.
- the line detected by the correction unit 141A may be a straight line (line segment) or a curved line.
- the detected line is not represented by multiple pixels.
- the detected line is specified at a position with decimal pixel accuracy. If the line segment is a line segment, it is a line segment connecting the start point and the end point, and if it is a curve, it is a smooth line connecting the start point and the end point.
- a well-known algorithm can be used by the correction unit 141A to detect a line consisting of continuous edges.
- FIG. 17 is a diagram for explaining an example of a process of correcting each frame by the correction unit 141A according to the second embodiment.
- FIG. 17A is a diagram showing another example of the region 561 in the first embodiment. In the region 561 in the second embodiment, the pixels are missing due to the large occurrence of aliases.
- the correction unit 141A detects the line 562 in the region 561 by processing in the same manner as the detection unit 142a of the first embodiment, as shown in FIG. 17 (b).
- the correction unit 141A calculates the average value of the pixel groups of the plurality of pixels through which the line 562 passes as the corrected pixel value.
- the correction unit 141A formulates the corrected pixel value I'i, where the i -th pixel value of the pixel group is I i , the corrected pixel value is I'i , and the number of pixels included in the pixel group is n. It can be calculated as shown in 18.
- FIG. 18 is a diagram for explaining an example of a process for correcting each frame according to the second embodiment.
- the correction unit 141A does not have to correct using the pixel values of all the pixels in the pixel group, samples one or more pixels, and corrects using the pixel values of the sampled one or more pixels. You may. For example, the frame may be corrected by replacing all the pixel values of the pixel groups of the plurality of pixels through which the line 562 passes by averaging the pixel values of one or more pixels sampled. Further, the correction unit 141A may use the line 562 to increase the resolution of the pixel group as shown in FIG. 18 (b).
- the restoration unit 142A performs matching to search for similarities between the frames corrected by the correction unit 141A, that is, a plurality of corrected frames, that is, the corrected first frame and the corrected second frame.
- the restoration unit 142A performs the same processing as the processing by the restoration unit 141 according to the first embodiment.
- step S104 the operation of the three-dimensional model generator 100A will be described. Since the three-dimensional model generation device 100A has different details of step S104 in the flowchart of FIG. 8 of the first embodiment, the details of step S104 will be described.
- FIG. 19 is a flowchart showing an example of details of the generation process of step S104 by the generation unit 140A according to the second embodiment.
- the generation unit 140A performs a loop 11 for each frame of the frame set of the multi-viewpoint images taken at the timings corresponding to each other (S121). In the loop 11, the processes of steps S122 and S123 are performed for each frame.
- the correction unit 141A detects the line of the subject in the frame by detecting the line consisting of continuous edges from the frame to be processed (S122).
- the correction unit 141A corrects the frame to be processed by using the line detected in the frame (S123).
- Loop 11 ends when steps S122 and S123 are completed for all frames.
- the restoration unit 142A processes a plurality of corrected frames, and performs a loop 12 for each first pixel of the plurality of first pixels of the first frame among the corrected plurality of frames (S124). ). In the loop 12, the processes of steps S125 and S126 are performed for each first pixel.
- the restoration unit 142A transfers the first pixel to be processed from the plurality of second pixels on the epipolar line corresponding to the first pixel on the plurality of second frames of the frame set to be processed to the first pixel. Search for similar similarities (S125).
- the restoration unit 142A generates a first three-dimensional point by performing a triangulation survey using the search result obtained in step S125. As a result, the restoration unit 142A generates a first three-dimensional model including a plurality of first three-dimensional points (S126).
- the loop 12 ends when the processing of steps S125 and S126 is completed for all the first pixels included in the first frame of the frame set to be processed.
- the three-dimensional model generation device 100A since the first three-dimensional model is generated using the image corrected by using lines, the error due to the alias on the image is reduced. You can generate a model.
- a plurality of frames obtained by photographing the subject from a plurality of viewpoints are acquired (S101), and one frame among the plurality of frames (first).
- a line consisting of continuous edges from the frame) is detected (S122), and a plurality of first three-dimensional points are generated based on the plurality of frames and the lines (S126). ).
- the three-dimensional model generation devices 100 and 100A exclude three-dimensional points whose distance from the viewpoint at which one frame is photographed is closer than a predetermined distance from the plurality of first three-dimensional points.
- the distance from the viewpoint of a plurality of third-dimensional points corresponding to the contours to which the lines correspond is larger than a predetermined distance based on the detected lines.
- the correction unit 142 does not have to correct the positions of the plurality of third-dimensional points by using the line. That is, the correction unit 142 does not have to correct the position of the second three-dimensional point located closer to the viewpoint than the predetermined distance among the plurality of generated second three-dimensional points. That is, the three-dimensional model generation device 100 may switch whether or not to perform correction for each line according to the distance from the viewpoint where one frame is captured to the contour of the subject to which the line corresponds.
- the processing load can be reduced without incurring a decrease in accuracy.
- the distance from the viewpoint of the subject when a distance from the viewpoint of the subject reflected in each frame can be obtained, the distance is closer to the viewpoint than a predetermined distance. It is not necessary to correct the pixel value of the pixel group in which the contour exists by using the line corresponding to the contour located at.
- the distance from the viewpoint of the subject may be estimated by using the three-dimensional model 520 obtained by the estimation process by the estimation device 200. That is, the three-dimensional model generation device 100 corrects the portion of the image corresponding to the contour according to the distance from the viewpoint where one frame is captured to the contour corresponding to the line for each line. You may switch whether or not.
- the three-dimensional model generation devices 100 and 100A may generate a plurality of first three-dimensional points without using edges detected from frames having a predetermined resolution or higher among the plurality of frames.
- the three-dimensional model generators 100 and 100A do not have to perform a process of detecting a line for a frame having a resolution equal to or higher than a predetermined resolution. This is because the accuracy of the first three-dimensional model generated by using a frame having a resolution equal to or higher than a predetermined resolution is higher than a certain level. That is, the three-dimensional model generation devices 100 and 100A may switch whether or not to detect a line for the frame according to the resolution of the frame.
- each processing unit included in the three-dimensional model generator or the like is realized by the CPU and the control program.
- each component of the processing unit may be composed of one or a plurality of electronic circuits.
- the one or more electronic circuits may be general-purpose circuits or dedicated circuits, respectively.
- the one or more electronic circuits may include, for example, a semiconductor device, an IC (Integrated Circuit), an LSI (Large Scale Integration), or the like.
- the IC or LSI may be integrated on one chip or may be integrated on a plurality of chips.
- IC integrated circuit
- LSI System LSI
- VLSI Very Large Scale Integration
- ULSI Ultra Large Scale Integration
- FPGA Field Programmable Gate Array
- the general or specific aspects of the present disclosure may be realized by a system, an apparatus, a method, an integrated circuit, or a computer program.
- a computer-readable non-temporary recording medium such as an optical disk, HDD (Hard Disk Drive) or semiconductor memory in which the computer program is stored.
- HDD Hard Disk Drive
- it may be realized by any combination of a system, an apparatus, a method, an integrated circuit, a computer program and a recording medium.
- This disclosure can be applied to a three-dimensional model generation device or a three-dimensional model generation system, and can be applied to, for example, figure creation, terrain or building structure recognition, person behavior recognition, free-viewpoint image generation, and the like.
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Computer Graphics (AREA)
- Geometry (AREA)
- Software Systems (AREA)
- Image Processing (AREA)
- Image Analysis (AREA)
- Apparatus For Radiation Diagnosis (AREA)
Abstract
Description
特許文献1に開示されている技術では、複数の画像間における類似点を探索することで三次元モデルを生成する。一般に、類似点の探索では、一の画像の一の画素の類似点を他の画像から探索する場合、カメラの幾何制約から他の画像上のエピポーラ線が算出され、エピポーラ線上の複数の画素について探索が行われる。そして、類似点の探索でマッチングされた画素と、画像を撮影したときのカメラの位置及び姿勢とを用いて、三角測量を行うことで、マッチングされた画素で撮影された三次元点の位置を特定する。これを複数の画素のそれぞれについて行うことで、複数の三次元点を含む三次元モデルを生成する。
[概要]
まず、図1を参照しながら、実施の形態1に係る三次元モデル生成方法の概要について説明する。
複数の撮像装置301は、所定の領域を撮影する複数の撮像装置である。複数の撮像装置301は、それぞれ被写体を撮影し、撮影した複数のフレームをそれぞれ推定装置200に出力する。本実施の形態では、撮像装置群300には、2台以上の撮像装置301が含まれる。また、複数の撮像装置301は、互いに異なる視点から同一の被写体を撮影する。フレームは、言い換えると、画像である。
推定装置200は、1台以上の撮像装置301に複数の視点から被写体を撮影させることでカメラ校正を行う。推定装置200は、例えば、複数の撮像装置301でそれぞれ撮影された複数のフレームに基づいて複数の撮像装置301の位置及び姿勢を推定するカメラ校正を行う。ここで、撮像装置301の姿勢とは、撮像装置301の撮影方向、及び、撮像装置301の傾きの少なくとも一方を示す。撮像装置301の撮影方向とは、撮像装置301の光軸の方向である。撮像装置301の傾きとは、基準姿勢からの撮像装置301の光軸周りの回転角度である。
三次元モデル生成装置100は、複数の撮像装置301で撮影された複数のフレームと推定装置200で推定されたカメラパラメータとに基づいて、所定の領域の三次元モデルを生成する。具体的には、三次元モデル生成装置100は、複数の撮像装置301それぞれに対応する複数のカメラパラメータと、複数のフレームとに基づいて、被写体の三次元モデルを仮想的な三次元空間上に生成する三次元モデル生成処理を実行する装置である。
次に、三次元モデル生成装置100の動作について、図8を用いて説明する。図8は、三次元モデル生成装置100の動作の一例を示すフローチャートである。
本実施の形態に係る三次元モデル生成方法は、被写体を第1視点から撮影することで得られた第1フレームと、被写体を第2視点から撮影することで得られた第2フレームとを取得し(S101)、少なくとも第1フレームにおいて、連続するエッジからなる線であって、被写体の輪郭に対応する線を検出し(S115)、第1フレーム、第2フレーム、および線とに基づいて、コンピュータ内の三次元空間において輪郭を表現する複数の第1三次元点を生成する(S116)。
(変形例1)
上記実施の形態1に係る三次元モデル生成方法では、三次元モデル生成装置100は、線で特定されたグループに属する複数の第3三次元点に対して主成分分析を行うことで、補正面を生成し、補正面を用いて複数の第3三次元点の位置を補正するとしたがこれに限らない。変形例1に係る三次元モデル生成方法では、複数の第3三次元点に対して最小二乗法を行うことで近似線を生成し、近似線を用いて複数の第3三次元点の位置を補正してもよい。
なお、法線は、式17に示すように、ベクトルの外積によって求めることができる。
変形例2について説明する。
実施の形態2に係る三次元モデル生成装置100Aについて説明する。
上記実施の形態1及び2では、被写体の輪郭に対応する線を用いて第1三次元モデルを生成する例について説明したが、全ての線を用いて第1三次元モデルを生成しなくてもよい。例えば、三次元モデル生成装置100、100Aは、1つのフレームが撮影された視点からの距離が所定の距離よりも近い位置にある三次元点を複数の第1三次元点から除く。
110 受信部
120 記憶部
130 取得部
140、140A 生成部
141、142A 復元部
142、141A 補正部
142a 検出部
142b 三次元点補正部
150 出力部
200 推定装置
300 撮像装置群
301 撮像装置
400 三次元モデル生成システム
500、510 被写体
520 三次元モデル
531 第1フレーム
532、533 第2フレーム
541 第1画素
542、543 第2画素
552、553 エピポーラ線
561、590 領域
562 線
570、580 第3三次元点
571 第1主成分
572 第2主成分
573 補正面
574、583 補正点
581 近似線
582 平面
L1 直線
D1、D2 距離
V1、V11 第1視点
V2、V12 第2視点
V3 第3視点
Claims (11)
- コンピュータによって実行される三次元モデル生成方法であって、
被写体を第1視点から撮影することで得られた第1画像と、前記被写体を第2視点から撮影することで得られた第2画像を取得し、
少なくとも前記第1画像において、連続するエッジからなる第1線であって、前記被写体の輪郭に対応する第1線を検出し、
前記第1画像、前記第2画像、および前記第1線に基づいて、前記コンピュータ内の三次元空間において前記輪郭を表現する複数の第1三次元点を生成する
三次元モデル生成方法。 - 前記第1線は、前記第1画像における画素単位より小さい単位で表現される
請求項1に記載の三次元モデル生成方法。 - さらに、
前記第1画像および前記第2画像において互いの類似点を探索するマッチングを行うことで、前記三次元空間において前記被写体を表現する複数の第2三次元点を生成し、
前記複数の第1三次元点の生成のために、
(i)前記第1線に基づいて、前記複数の第2三次元点のうち、前記輪郭に対応する複数の第3三次元点を特定し、
(ii)前記複数の第3三次元点の位置を補正する
請求項1又は2に記載の三次元モデル生成方法。 - さらに、
前記第2画像において、連続するエッジからなる第2線であって、前記第1線が対応する輪郭の少なくとも一部を含む輪郭に対応する第2線を検出し、
前記複数の第1三次元点の生成のために、さらに、
(i)前記第2線に基づいて、前記複数の第2三次元点のうち、前記第2線が対応する輪郭に対応する複数の第4三次元点であって、補正された前記複数の第3三次元点と補正されていない第2三次元点とを含む複数の第4三次元点を特定し、
(ii)前記複数の第4三次元点のうち、少なくとも前記補正されていない第2三次元点の位置を補正する
請求項3に記載の三次元モデル生成方法。 - 前記複数の第1三次元点の生成のために、前記複数の第3三次元点に対して主成分分析を行うことで平面を特定し、前記複数の第3三次元点が前記平面に近づくように前記複数の第3三次元点の位置を補正する
請求項3又は4に記載の三次元モデル生成方法。 - 前記複数の第1三次元点の生成のために、前記複数の第3三次元点の重心を通過し、かつ、前記主成分分析で得られた第1主成分を法線とする第1平面、及び、前記重心を通過し、かつ、前記主成分分析で得られた第2主成分を法線とする第2平面のうちの、前記複数の第3三次元点からの距離の総和が小さい平面を前記平面として生成する
請求項5に記載の三次元モデル生成方法。 - 前記第1三次元点の生成のために、前記複数の第3三次元点に対して最小二乗法を行うことで近似線を生成し、前記複数の第3三次元点が前記近似線に近づくように前記複数の第3三次元点の位置を補正する
請求項3又は4に記載の三次元モデル生成方法。 - 前記複数の第1三次元点の生成のために、前記第1視点からの距離が所定の距離より近い位置にある三次元点を前記複数の第1三次元点から除く
請求項1から7のいずれか1項に記載の三次元モデル生成方法。 - 前記複数の第1三次元点の生成のために、前記第1画像および前記第2画像のうちで所定の解像度以上の画像から検出されたエッジを用いない
請求項1から8のいずれか1項に記載の三次元モデル生成方法。 - 前記検出では、前記第2画像において、連続するエッジからなる第2線であって、前記被写体の輪郭に対応する第2線を検出し、
前記複数の第1三次元点の生成では、(i)前記第1画像および前記第2画像のそれぞれを、前記第1線および前記第2線を用いて補正し、(ii)補正後の前記第1画像および補正後の前記第2画像において互いの類似点を探索するマッチングを行う
請求項1又は2に記載の三次元モデル生成方法。 - コンピュータを備え、
前記コンピュータは、
被写体を第1視点から撮影することで得られた第1画像と、前記被写体を第2始点から撮影することで得られた第2画像を取得し、
少なくとも前記第1画像において、連続するエッジからなる第1線であって、前記被写体の輪郭に対応する第1線を検出し、
前記第1画像、前記第2画像、および前記第1線に基づいて、前記コンピュータ内の三次元空間において前記輪郭を表現する複数の第1三次元点を生成する
三次元モデル生成装置。
Priority Applications (4)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202180081018.7A CN116547493A (zh) | 2020-12-08 | 2021-11-15 | 三维模型生成方法及三维模型生成装置 |
EP21903121.8A EP4261496A4 (en) | 2020-12-08 | 2021-11-15 | METHOD AND DEVICE FOR CREATING A THREE-DIMENSIONAL MODEL |
JP2022568137A JPWO2022124017A1 (ja) | 2020-12-08 | 2021-11-15 | |
US18/203,925 US20230394754A1 (en) | 2020-12-08 | 2023-05-31 | Three-dimensional model generation method, three-dimensional model generation device, and non-transitory computer readable medium |
Applications Claiming Priority (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
JP2020203334 | 2020-12-08 | ||
JP2020-203334 | 2020-12-08 |
Related Child Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
US18/203,925 Continuation US20230394754A1 (en) | 2020-12-08 | 2023-05-31 | Three-dimensional model generation method, three-dimensional model generation device, and non-transitory computer readable medium |
Publications (1)
Publication Number | Publication Date |
---|---|
WO2022124017A1 true WO2022124017A1 (ja) | 2022-06-16 |
Family
ID=81974374
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
PCT/JP2021/041943 WO2022124017A1 (ja) | 2020-12-08 | 2021-11-15 | 三次元モデル生成方法及び三次元モデル生成装置 |
Country Status (6)
Country | Link |
---|---|
US (1) | US20230394754A1 (ja) |
EP (1) | EP4261496A4 (ja) |
JP (1) | JPWO2022124017A1 (ja) |
CN (1) | CN116547493A (ja) |
TW (1) | TW202230288A (ja) |
WO (1) | WO2022124017A1 (ja) |
Citations (9)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2011053809A (ja) * | 2009-08-31 | 2011-03-17 | Fuji Heavy Ind Ltd | 車両用白線認識装置 |
JP2011100174A (ja) * | 2009-11-03 | 2011-05-19 | Tokyo Institute Of Technology | 車線内車両検出装置及び車線内車両検出方法 |
JP2013068417A (ja) * | 2011-09-20 | 2013-04-18 | Dainippon Screen Mfg Co Ltd | 三次元位置・姿勢認識装置、産業用ロボット、三次元位置・姿勢認識方法、プログラム、記録媒体 |
JP2013250907A (ja) * | 2012-06-04 | 2013-12-12 | Ricoh Co Ltd | 視差算出装置、視差算出方法及び視差算出用プログラム |
JP2014235750A (ja) * | 2013-06-03 | 2014-12-15 | 株式会社リコー | 目標検出方法及び目標検出装置 |
JP2016014610A (ja) * | 2014-07-02 | 2016-01-28 | 株式会社リコー | カメラシステム、測距方法、およびプログラム |
JP2017130146A (ja) | 2016-01-22 | 2017-07-27 | キヤノン株式会社 | 画像管理装置、画像管理方法及びプログラム |
JP2020047049A (ja) * | 2018-09-20 | 2020-03-26 | ファナック株式会社 | 画像処理装置及び画像処理方法 |
JP2020071034A (ja) * | 2018-10-29 | 2020-05-07 | セイコーエプソン株式会社 | 三次元計測方法、三次元計測装置、及び、ロボットシステム |
Family Cites Families (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP5746477B2 (ja) * | 2010-02-26 | 2015-07-08 | キヤノン株式会社 | モデル生成装置、3次元計測装置、それらの制御方法及びプログラム |
US9846963B2 (en) * | 2014-10-03 | 2017-12-19 | Samsung Electronics Co., Ltd. | 3-dimensional model generation using edges |
-
2021
- 2021-11-15 CN CN202180081018.7A patent/CN116547493A/zh active Pending
- 2021-11-15 WO PCT/JP2021/041943 patent/WO2022124017A1/ja active Application Filing
- 2021-11-15 JP JP2022568137A patent/JPWO2022124017A1/ja active Pending
- 2021-11-15 EP EP21903121.8A patent/EP4261496A4/en active Pending
- 2021-11-24 TW TW110143825A patent/TW202230288A/zh unknown
-
2023
- 2023-05-31 US US18/203,925 patent/US20230394754A1/en active Pending
Patent Citations (9)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2011053809A (ja) * | 2009-08-31 | 2011-03-17 | Fuji Heavy Ind Ltd | 車両用白線認識装置 |
JP2011100174A (ja) * | 2009-11-03 | 2011-05-19 | Tokyo Institute Of Technology | 車線内車両検出装置及び車線内車両検出方法 |
JP2013068417A (ja) * | 2011-09-20 | 2013-04-18 | Dainippon Screen Mfg Co Ltd | 三次元位置・姿勢認識装置、産業用ロボット、三次元位置・姿勢認識方法、プログラム、記録媒体 |
JP2013250907A (ja) * | 2012-06-04 | 2013-12-12 | Ricoh Co Ltd | 視差算出装置、視差算出方法及び視差算出用プログラム |
JP2014235750A (ja) * | 2013-06-03 | 2014-12-15 | 株式会社リコー | 目標検出方法及び目標検出装置 |
JP2016014610A (ja) * | 2014-07-02 | 2016-01-28 | 株式会社リコー | カメラシステム、測距方法、およびプログラム |
JP2017130146A (ja) | 2016-01-22 | 2017-07-27 | キヤノン株式会社 | 画像管理装置、画像管理方法及びプログラム |
JP2020047049A (ja) * | 2018-09-20 | 2020-03-26 | ファナック株式会社 | 画像処理装置及び画像処理方法 |
JP2020071034A (ja) * | 2018-10-29 | 2020-05-07 | セイコーエプソン株式会社 | 三次元計測方法、三次元計測装置、及び、ロボットシステム |
Non-Patent Citations (2)
Title |
---|
See also references of EP4261496A4 |
UKIDA, HIROYUKI: "3D Shape Reconstruction of Colored Object Using Image Scanner", PAPERS OF TECHNICAL MEETING IEE JAPAN (IEEJ), no. IP-04-16, 17 September 2004 (2004-09-17), pages 37 - 44, XP009537555 * |
Also Published As
Publication number | Publication date |
---|---|
JPWO2022124017A1 (ja) | 2022-06-16 |
US20230394754A1 (en) | 2023-12-07 |
CN116547493A (zh) | 2023-08-04 |
TW202230288A (zh) | 2022-08-01 |
EP4261496A4 (en) | 2024-05-29 |
EP4261496A1 (en) | 2023-10-18 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
US11830216B2 (en) | Information processing apparatus, information processing method, and storage medium | |
WO2011013301A1 (en) | Position and orientation calibration method and apparatus | |
WO2021140886A1 (ja) | 三次元モデル生成方法、情報処理装置およびプログラム | |
WO2018171008A1 (zh) | 一种基于光场图像的高光区域修复方法 | |
JP2011174879A (ja) | 位置姿勢推定装置及びその方法 | |
CN107560603B (zh) | 一种无人机倾斜摄影测量系统及测量方法 | |
CN103971378A (zh) | 一种混合视觉系统中全景图像的三维重建方法 | |
JP7479324B2 (ja) | 情報処理装置、情報処理方法、及びプログラム | |
CN112184811B (zh) | 单目空间结构光系统结构校准方法及装置 | |
US10607350B2 (en) | Method of detecting and describing features from an intensity image | |
US20200143603A1 (en) | Information processing apparatus, information processing method, and non-transitory computer-readable storage medium | |
CN114494462A (zh) | 一种基于Yolov5和改进跟踪算法的双目相机测距方法 | |
JP6061770B2 (ja) | カメラ姿勢推定装置及びそのプログラム | |
CN111882655A (zh) | 三维重建的方法、装置、系统、计算机设备和存储介质 | |
WO2016135856A1 (ja) | 3次元形状計測システムおよびその計測方法 | |
US20230005216A1 (en) | Three-dimensional model generation method and three-dimensional model generation device | |
JP6922348B2 (ja) | 情報処理装置、方法、及びプログラム | |
JP7498404B2 (ja) | 被写体の3次元姿勢推定装置、3次元姿勢推定方法、及びプログラム | |
WO2022124017A1 (ja) | 三次元モデル生成方法及び三次元モデル生成装置 | |
Li et al. | Joint intrinsic and extrinsic lidar-camera calibration in targetless environments using plane-constrained bundle adjustment | |
JP3221384B2 (ja) | 三次元座標計測装置 | |
WO2019058487A1 (ja) | 3次元復元画像処理装置、3次元復元画像処理方法及び3次元復元画像処理プログラムを記憶したコンピュータ読み取り可能な記憶媒体 | |
CN116704111B (zh) | 图像处理方法和设备 | |
WO2023095375A1 (ja) | 三次元モデル生成方法及び三次元モデル生成装置 | |
Paudel et al. | Localization of 2D cameras in a known environment using direct 2D-3D registration |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
121 | Ep: the epo has been informed by wipo that ep was designated in this application |
Ref document number: 21903121 Country of ref document: EP Kind code of ref document: A1 |
|
ENP | Entry into the national phase |
Ref document number: 2022568137 Country of ref document: JP Kind code of ref document: A |
|
WWE | Wipo information: entry into national phase |
Ref document number: 202180081018.7 Country of ref document: CN |
|
NENP | Non-entry into the national phase |
Ref country code: DE |
|
ENP | Entry into the national phase |
Ref document number: 2021903121 Country of ref document: EP Effective date: 20230710 |